Extraction of Gold

There are four main processes of gold extraction:

1. Mechanical processes for preparing and washing the ore.
2. Preparation of the ores, with simultaneous or subsequent amalgamation.
3. Chemical extraction-processes.
4. Smelting of certain ores.

Gold can be extracted by washing only when it is present as metal in the form of particles which are not too minute. The washing removes the specifically lighter parts of the material, the heavier gold particles sinking to the bottom. The process is applicable directly to gold- bearing sand, and to gold-bearing rocks after crushing. The crushing is effected by stone-crushers, rolling-mills, and stamp-mills. The washing is carried out either by subjecting the material to the prolonged action of flowing water, or by mechanical agitation with water in deep iron pans or " cradles." In California hydraulic jets are employed to disintegrate the material, the water being allowed to flow into long channels, and the gold extracted by amalgamation with mercury.

With pyritic ores a preliminary roasting to eliminate sulphur is necessary. The ore is reduced to a fine state of division and amalgamated simultaneously in a stamp-mill, the solid gold-amalgam separated from the liquid mercury by filtration through leather under pressure, and the mercury distilled. Complete extraction of the gold by amalgamation is impossible, a part remaining in the mud of the stamp- mills. Concentration is effected by washing, and the " concentrates " are chlorinated by the Plattner process. A further operation divides the waste products into 60 per cent, of " tailings " and 40 per cent, of "slimes," and these materials are further extracted by the cyanide-process.

Plattner's Chlorination-process

As a preliminary step the material is submitted to an oxidizing and chlorinating roasting, to convert metals other than gold into chlorides. After chlorination of the gold, preferably with chlorine under pressure, the metal is precipitated by hydrogen sulphide or ferrous sulphate, or the gold is precipitated by filtration through charcoal, and recovered by combustion of the charcoal. Direct generation of the chlorine in the liquid has also been suggested. It can be effected by addition of bleaching-powder and sulphuric acid; or by means of manganese dioxide, sodium chloride, and sulphuric acid; or electrolytically. Bromine acts more energetically than chlorine, and has also been employed in the extraction. The recovery of the gold is effected similarly, and the bromine is then liberated by the action of chlorine, but it is impossible to prevent loss of bromine.

Cyanide-process

The extraction of gold by the cyanide-method is of great technical importance. It is effected by lixiviation with a solution of potassium cyanide, the gold being precipitated either by addition of zinc or electrolytically. The process was suggested by Mac Arthur and Forrest in 1385, and is based on the well-known solubility of gold in potassium-cyanide solution, a phenomenon said to have been discovered by Prince Bagration in 1843. It was first employed on the large scale in the Australian goldfields in 1888.

The removal of metallic salts, such as ferrous sulphate, and acids from the material being handled is effected by lixiviation with dilute alkali. The residue is extracted with 0.35 per cent, solution of potassium cyanide, then with 0.08 per cent, solution, and ultimately with water. The extract is transferred to tanks, and the gold precipitated as a powder by addition of zinc. It is then washed out from the bottom of the tanks, dried, and freed from zinc by roasting and fusing.

In the electrolytic precipitation of the gold an iron anode and a sheet-lead cathode are employed, the current-density being very low, about 0.5 amp. per sq. metre. The gold is deposited on the lead, and after removal of this metal still contains a considerable proportion of both lead and silver. It is freed from them by the operation called " parting."

Access of air is essential to solution of gold in potassium cyanide, the process being attended by evolution of hydrogen. Lead, bismuth, antimony, cadmium, silver, and mercury also dissolve in presence of air; but copper, iron, aluminium, nickel, cobalt, and zinc dissolve in absence of air. Gold and silver are distinguished by the fact that their maximum solubility corresponds with a very low concentration of the potassium-cyanide solution, a phenomenon probably due to the slight solubility of air in concentrated solutions of this salt. The solution of gold in the cyanide solution is accompanied by the intermediate formation of hydrogen peroxide, and the process is accelerated by addition of this substance:

2Au+4KCN+2H₂O+O₂ = 2KAu(CN)₂+2KOH+H₂O₂;
2Au+4KCN+H₂O₂ = 2KAu(CN)₂+2KOH.

A similar accelerating effect is exerted by other substances, such as potassium ferricyanide, potassium permanganate, potassium chromate, sodium peroxide, barium peroxide, cyanogen bromide, cyanogen chloride, persulphates, and certain organic compounds. The best method of reducing the proportion of the other metals is to maintain the cyanide solution dilute.

In precipitating the gold by zinc, the proportion required is about seven times that indicated by the equation

Zn+2Au^•=2Au + Zn^••,

the discrepancy being due to solution of part of the zinc in the cyanide solution, with evolution of hydrogen. Purity of the zinc is an important factor in counteracting this loss.

In the electrolytic deposition of gold from cyanide solutions hydrogen is liberated at the cathode, and an equivalent number of hydroxyl ions give up their charges at the anode, the solution developing an alkaline reaction. Cyanogen ions also give up their charges at the anode, being probably converted partly into cyanate, and partly into complex derivatives of iron and cyanogen, such as Prussian blue. The formation of this product is prevented by replacement of the iron anode by one made of lead peroxide.

In the Pelatan-Clerici process the gold is dissolved electrolytically and precipitated in a single operation. Cathodes of mercury or amalgamated copper are employed, so that large particles of gold which fail to dissolve are directly amalgamated.

The smelting process consists in the formation of an alloy of gold with silver and lead, and is similar to that employed in extracting silver. It is applicable to ores rich in silver, and also to refractory ores containing arsenic and antimony, for which the other processes are unsuitable.

Parting of Gold from Silver

The parting of gold from silver is effected by melting the raw product with sufficient silver to make the ratio of the gold to silver 1:3, or " quartering," the alloy being then treated with concentrated sulphuric acid or nitric acid. Another method of parting consists in transforming the silver present in the alloy into chloride, as in the aqua-regia and electrolytic processes.

Quartering

This designation is applied to the parting with nitric acid mentioned above, although it is also employed to describe the sulphuric-acid process. Usually, the gold-silver alloy is made to contain 2 parts of gold to 5 parts of silver. It is granulated, and boiled with nitric acid. The silver solution produced is either worked up into " lunar caustic," or precipitated as chloride by addition of sodium chloride, and then reduced to metal.

Refining

The gold is alloyed with silver and granulated, as in quartering. It is then boiled with concentrated sulphuric acid in cast- iron boilers, silver, copper, lead, and other metals being converted into sulphates. The silver and lead salts are partly dissolved in the strong acid, but copper sulphate remains chiefly undissolved, and by coating the granules exercises a retarding influence on the solution of the silver, thus necessitating a repetition of the boiling with acid. All the gold is in the undissolved residue, and after washing thoroughly it is dried and fused. It still contains traces of silver, but is suitable for most purposes without further purification. The silver solution is diluted with water, the metal precipitated by the action of copper, and the solution formed worked up for cupric sulphate. Other processes are reduction of the silver sulphate by ferrous sulphate; and by iron, which precipitates silver only, but no copper.

When the gold contains platinum and related metals, it is submitted to further refining. Part of the platinum is dissolved out by the treatment with nitric acid. Iridium was formerly slagged off by fusion with potassium nitrate, but all the platinum is simultaneously slagged, and can be recovered only by a cumbrous process. To obviate this difficulty the metal, either after parting from silver or in its original state, is dissolved, and the gold and platinum precipitated separately.

Parting by aqua regia

Parting by aqua regia is effected by dissolving the alloyed gold in the acid, silver being converted into its chloride, which is then precipitated by dilution with water. The platinum dissolves completely, and the iridium partially, the gold being precipitated by addition of ferrous sulphate or chloride. With a large proportion of silver some of the gold is occluded, and escapes solution in the acid, thus necessitating a repetition of the treatment with acid. The process lacks many of the advantages characteristic of the electrolytic method.

The electrolytic process was introduced in 1863 by Charles Watt at Sydney, started in 1878 by Wohlwill at Hamburg, and in 1902 by Tuttle at the Philadelphia Mint. In the gold-chloride method the electrolyte is a solution of auric chloride containing free hydrochloric acid, the crude metal forming the anode, and pure sheet gold the cathode. The gold dissolved at the anode is deposited in a pure condition at the cathode. Other metals are converted into chlorides at the anode, and either remain dissolved or pass into the anodic slime. Silver is converted into its chloride, this substance partly dissolving, partly depositing in the slime, and partly adhering to the anode. With solutions containing more than 3 to 10 per cent, of hydrochloric acid, and with bullion having more than 6 per cent, of silver, the coating of the anode raises the density of the current and causes evolution of chlorine.

Rose has found that with an electrolyte containing 29 per cent, of free hydrochloric acid, and with a current-density of 5000 amperes per square metre of anode surface, no chlorine is evolved, even with an anode containing 20 per cent, of silver. The heavy current causes the silver chloride to separate from the anode, and as aurous chloride is not allowed to form, deposition of gold in the anode-slime is prevented.

A solution of auric chloride containing 3 to 5 per cent, of gold and a current-density of 1000 amperes per square metre are usually employed, but Rose has found that with a current-density of 5000 amperes per square metre an electrolyte with 20 per cent, of gold yields a coherent deposit capable of being readily washed, and malleable after melting. By this modification the time required for solution of the anode is reduced from one week to one day.

In addition to the malleable nature of the product obtained by electrolytic refining, the process also extracts platinum, a constituent of nearly all samples of Transvaal gold. The United States of America Mint has found the electrolytic method more economical than that with sulphuric acid.

Miller's dry parting process involves the action of chlorine on molten gold covered with a layer of borax to prevent spurting. The gold is not attacked, but the silver is converted into chloride. When the gold has solidified, the molten silver chloride collected on the surface is run off, carrying with it a small proportion of gold. When silver is the chief impurity in the gold, as in that found in Australia, the process is specially applicable. The gold has a degree of purity of 99.1 to 99.7 per cent.

The metal is known in several modifications. In the crystallized state it belongs to the cubic system, a great variety of forms having been observed. Hexagonal forms have also been described. Artificial crystals can be prepared by heating 5 per cent, gold-amalgam at 80° C., and acting on the product with nitric acid of density 1.35. References are appended to other modes of producing crystals. Rolled, non-crystalline gold becomes crystalline by exposure to red heat. This process is reversed by polishing or hammering, with formation of a superficial layer of amorphous gold.

Reducers precipitate gold from different solutions in divergent forms. Ferrous chloride and sulphate, arsenious acid, antimonious acid, and stannous chloride throw it down as a brown powder of varying degrees of subdivision, the precipitate with ferrous chloride being more finely divided when the gold solution is poured into the iron solution than that produced by the reverse method. The more dilute the gold solution, the finer is the subdivision of the precipitate. From concentrated solutions the metal often separates in lustrous laminae. A soft, yellow gold sponge is produced by addition of a small proportion of oxalic acid and a large proportion of potassium carbonate to a concentrated solution, the resulting mixture being then boiled with more oxalic acid.

The so-called " brown gold " is formed by the action of nitric acid on an alloy of gold and silver containing 20 per cent, of gold. The silver dissolves, leaving the gold as a brown, spongy sheet, reconverted at 1040° C. into ordinary gold. The brown form (β) differs from ordinary (α) gold in density and magnetic properties. Above 700° C. it undergoes slow transformation into the ordinary variety. It is uncertain whether the difference is caused by dimorphism or by allotropy.

Thomsen believed that he had demonstrated the existence of three allotropic modifications of gold, relying on thermochemical measurements, but in opposition to his conclusions is the fact that there is no potential difference between his supposed modifications.

With very dilute solutions most precipitants yield a turbidity or coloration, gold being only slowly deposited as a brown powder. Various organic reducers give intensely coloured red or blue solutions of colloidal gold, also prepared by Faraday by the action of white phosphorus on auric-chloride solution, his product being a red liquid from which metallic gold gradually deposited. Other reducers available for the preparation of colloidal gold solutions are oxalic acid, sulphurous acid, glycerol, sodium hypophosphite, formaldehyde, hydrazine hydrate, acetylene, phenylhydrazine hydrochloride, sodium hyposulphite, a mixture of carbon monoxide and dioxide, an alcoholic solution of phosphorus, alcohol, oil of turpentine, acraldehyde, allyl alcohol, polyhydric phenols, aluminium, humic acid, starch, and hydrogen peroxide. Adrenaline, alloxan, tannic acid, and p-phenylenedimethyldiamine have also been employed.

In using sodium hypophosphite, Carey Lea mixed a 10 per cent, solution of the reagent with 1 c.c. of auric-chloride solution containing 0.1 gram of gold and a drop of sulphuric acid, and added 30 c.c. of water as soon as darkening of the solution had begun. A gradual precipitation of bluish-black metallic gold supervened, and the precipitate was removed by filtration, but the green filtrate gradually became turbid, and deposited gold on the walls of the container. By reflected light the precipitated metal was yellowish brown in colour; by transmitted light it was bright blue.

Henriot employed catechol in presence of sodium carbonate as a reducer.

Sunlight greatly facilitates the formation of colloidal gold, but the light of a mercury lamp is less energetic. Certain soaps are said to exert a very pronounced protective action in the formation of gold hydrosols.

A summary of investigations on colloidal gold has been given by Ostwald, and also by Cornejo. Its optical properties have been examined by Steubing, and the various colours of the product have been studied by Bancroft.

Purple of Cassius, first prepared by Andreas Cassius at Leyden in 1683, is produced as a dark-brown precipitate, purple-red by transmitted light, by the action of stannous salts on dilute gold solutions. Its preparation is effected by boiling an aqueous solution of stannous chloride with freshly precipitated ferric hydroxide, dissolving the precipitated stannous hydroxide in hydrochloric acid, and adding the solution drop by drop to a very dilute gold solution. Metallic tin or stannous sulphate or nitrate can be substituted for stannous chloride. Another technical method for its preparation is to heat "pink salt," SnCl₄,2NH₄Cl, with tin-foil and water until solution of the tin is complete, dilute with water, and pour the solution into a dilute aqueous gold solution. It is also produced by dissolving gold-tin alloys in nitric acid. The percentage of gold in purple of Cassius varies between 24 and 43 per cent, with the experimental conditions of its preparation. The substance contains 7 to 14 per cent, of water, a part of this water being expelled at 100° C., and the remainder at red heat. As " gold lake " it finds technical application in colouring glass ruby-red, and also in colouring enamels and glazes.

There has been much diversity of opinion as to the constitution of purple of Cassius. Macquer regarded it as a mixture of gold and hydrated oxide of tin, since increase in the amount of gold changes the colour of the solution to a darker purple. This view is apparently in opposition to Proust's observation that moist purple of Cassius is soluble in ammonium hydroxide, but trituration with mercury fails to effect separation of the gold. Proust regarded it as a mixture of auro-stannous stannate and stannic hydroxide. Debray compared it to the lakes, suggesting it to be stannic and metastannic acids coloured by finely divided gold, the metal being rendered insoluble in mercury by a process analogous to that undergone by dyes during deposition on the fibre. This view is supported by the possibility of depositing gold from ammoniacal solutions of purple of Cassius under the influence of light, since there is no method known for precipitating gold from its oxides by means of ammonia. Zsigmondy regarded purple of Cassius as a mixture of colloidal gold and colloidal stannic acid, a view practically identical with that of Debray expressed in modern phraseology.